INDEX

Explanations

words related to general evaluations or summaries of a text

oai_token-act-pair · gpt-3.5-turbo

general evaluations or overall impressions

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.1.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Typh

-0.71

ret

-0.70

href

-0.68

fle

-0.67

lev

-0.66

 Frie

-0.64

bub

-0.64

 Ferdinand

-0.64

lets

-0.63

 MÃ©

-0.63

POSITIVE LOGITS

Overall

1.12

 Overall

1.12

Rating

0.90

gebra

0.77

 constitu

0.77

 Defin

0.76

 Satisf

0.74

 Length

0.72

 rating

0.71

 conclud

0.71

Activations Density 0.012%

words related to general evaluations or summaries of a text

general evaluations or overall impressions

No Comments

No Known Activations